The maternal nutrient environment influences pregnancy outcomes and fetal growth, and there is evidence of its impact on postnatal outcomes that can persist into childhood and adulthood. Over 2/3 of the pregnancies of HIV-infected women have complications. There is also concern that HIV-exposed, but uninfected children can have problems with growth and neurocognition. These complications are likely multifactorial and may be related to ART and substance abuse. However, there are limited comprehensive data evaluating maternal diets of HIV-infected women with particular reference to dietary patterns, food groups and specific micro and macronutrients known to impact on pregnancy, perinatal, and infant outcomes. This study, will take advantage of the NICHD-funded Pediatric HIV/AIDS Cohort Study (PHACS) infrastructure to evaluate dietary parameters and associations with maternal and infant outcomes. We will specifically evaluate the dairy and fish food groups as well as specific polyunsaturated fat intakes (n-3, n-6 and ratios), iron and vitamin D. Our main objectives are to: 1. characterize dietary patterns, food groups and nutrient intake in the 3rd trimester of pregnancy in HIV-infected women;2. determine the associations between these dietary factors in pregnancy and pregnancy, perinatal, and infant outcomes and;3. determine the specific associations of maternal biomarkers of iron and vitamin D, a marker of inflammation (hs-CRP) on our outcomes of interest. We will recruit 230 women (with goal of analyzing 200) who are enrolled in the SMARTT protocol over the next 2 years. We will obtain 3 telephone administered 24 hour diet recalls during their 3rd trimester of pregnancy and obtain a blood sample for 25-0H vitamin D, hs-CRP, iron studies, and calcium (and will bank blood). We will record pregnancy and perinatal complications (per the PHACS protocol) and will obtain 2-24 hour diet recalls, neurocognitive development and growth assessments through the 1 st year of life (not all children will reach 1 yr during this 2-year grant interval;but a sufficient number [20%] will have analyzable 1 year outcomes for this specific endpoint). With this information, we Will be able to determine associations between maternal diet and 1. pregnancy, 2. perinatal and 3. early infant outcomes in women with HIV so that appropriate nutritional recommendations can be made for HIV-infected women in pregnancy.